In this project, atomic absorption spectroscopy, nuclear magnetic resonance spectroscopy, and ion selective electrodes are being used to characterize the relative fractions of magnesium in the ionized, protein-bound, and ligand-complexed pools of the blood for normal and sick preterm and term neonates as well as maternal blood and cord blood. The broad goal is to use this new technology to obtain a more complete characterization of the distribution of Mg in blood, which in turn will provide a better understanding of magnesium regulation in pregnant women, preterm and sick infants, and lead to useful new clinical information which may alter the present management of these patients. Our preliminary results demonstrate the feasibility of quantifying the Mg pools in plasma, and the potential for large variations in the size of these pools to occur in both mother and infant during and following MgSO4 therapy. The data demonstrate that the intravenous infusion of MgSO4 does not simply add Mg2+ ions to a static solution of plasma. Although it has been recognized that the kidney is the main regulatory organ of Mg homeostasis, the preliminary results suggest that plasma is capable of partially buffering acute increases in ionized Mg, probably due to a combination of Mg buffering from both high and low affinity Mg binding ligands present in blood. There are indications that the extent of this buffering increases in response to hypermagnesemia over a period of hours. (Collaborative 4) REPORT PERIOD: (09/01/97-08/31/98)